Overview

Summary: This report explores adult fish passage recorded from 2001-01-01 to 2010-12-31 at the Willamette Falls fish ladder on the Willamette River, Oregon. This report is focused on Coho, Jack Coho, and Steelhead species. First, a time series graph visualized time series data by month and year to explore trends over the 10-year time period. Second, a season plot is created to explore seasonal trends for each species. Lastly, a total counts figure illustrates the total number of observed fish passage for each species over the 10-year period.

Image of Willamette Falls, OR:

knitr::include_graphics('willamette_falls.jpeg')
Photo of Willamette Falls, Oregon. Credit: EncMstr, CC BY-SA 3.0 <http://creativecommons.org/licenses/by-sa/3.0/>, via Wikimedia Commons

Photo of Willamette Falls, Oregon. Credit: EncMstr, CC BY-SA 3.0 http://creativecommons.org/licenses/by-sa/3.0/, via Wikimedia Commons

Map of Willamette Falls, OR:

knitr::include_graphics('willamette_falls_map.png')
Map of the location of the fish ladder in Willamette Falls, Oregon. Credit: Google Maps

Map of the location of the fish ladder in Willamette Falls, Oregon. Credit: Google Maps

Data Source: University of Washington. (n.d.). DART Adult Passage Graphics & Text. Columbia Basin Research. http://www.cbr.washington.edu/dart/query/adult_graph_text

Adult Fish Passage at Willamette Falls

Time series

## Read in packages 
library(tidyverse)
library(here)
library(janitor)
library(lubridate)
library(tsibble)

## Read in the data, keep only the columns of interest, replace "NA" with 0, and clean names
fish_af <- read.csv(here("data", "willamette_fish_passage.csv")) %>% 
  select("Date", "Coho", "Jack.Coho", "Steelhead") %>% 
  mutate_all(~replace(., is.na(.), 0)) %>% 
  clean_names()
  
## Convert to tsibble 
fish_ts <- fish_af %>% 
  mutate(date = lubridate::mdy(date)) %>% 
  as_tsibble(key = NULL, index = date)

## Plot the time series data
# ggplot() +
#   geom_line(data = fish_ts, aes(x = date, y = coho)) +
#   labs(x = "Date",
#        y = "Number of Coho")

## Make data frames for each fish by monthyears  
coho_month <- fish_ts %>% 
  index_by(yr_mo = ~yearmonth(.)) %>% 
  summarize(monthly_mean_coho = mean(coho, na.rm = TRUE))

jack_month <- fish_ts %>% 
  index_by(yr_mo = ~yearmonth(.)) %>% 
  summarize(monthly_mean_jack = mean(jack_coho, na.rm = TRUE))

steel_month <- fish_ts %>% 
  index_by(yr_mo = ~yearmonth(.)) %>% 
  summarize(monthly_mean_steel = mean(steelhead, na.rm = TRUE))
## Plot time series fish passage for the three species with month/year data 
ggplot() +
  geom_line(data = coho_month,
       aes(x = yr_mo,
           y = monthly_mean_coho,
           colour = "Coho"),
        alpha = 0.8,
        size = 1) +
  geom_line(data = jack_month,
       aes(x = yr_mo,
           y = monthly_mean_jack,
           colour = "Jack Coho"),
       alpha = 0.8,
       size = 1) +
  geom_line(data = steel_month,
       aes(x = yr_mo,
           y = monthly_mean_steel,
           colour = "Steelhead"),
       alpha = 0.8,
      size = 1) +
  labs(x = "Month and Year",
       y = "Number of Fish Observed",
       colour = "Species",
       title = "Adult Passage of Coho, Jack Coho, and Steelhead at Willamette Falls") +
  theme_minimal() +
  scale_colour_manual(values = c("dodgerblue4", "darkorchid4", "springgreen4"))

Figure 1.1 Time series plot of Coho, Jack Coho, and Steelhead adult fish passage at the Willamette Falls fish ladder between 2001 and 2010, plotting the average fish numbers for each month. Data were shared by and accessed from Columbia River DART (Data Access in Real Time).

Summary

  • Between 2001 and 2010, the number of Coho observed increased over time, reaching the highest count in the last year of observations.
  • Jack Coho observed counts are low through all the years of observation in comparison with Steelhead and some years of Coho.
  • There is seasonality in the data; we see high and low points each year for each species.
  • It appears that Coho Salmon pass through the fish ladder later in the year than Steelhead Trout, and around the same time as Jack Coho.

Seasonplots

# Attach feasts package
library(feasts)
# Pivot longer
fish_ts_longer <- fish_ts %>%
  pivot_longer(2:4,
               names_to = "species", # Move each species to its own column
               values_to = "n") %>%
  group_by(species) %>% # Group by species
  mutate(species = case_when(
    species == "coho" ~ "Coho", 
    species == "jack_coho" ~ "Jack Coho", 
    species == "steelhead" ~ "Steelhead")
  ) %>%
  index_by(yr_mo = ~yearmonth(.)) %>% # Group by year & month
  summarize(monthly_mean = mean(n, na.rm = TRUE)) # Summarize the mean for each species for each month

# Make vector for species color
fish_color = c("Coho" = "dodgerblue4", "Jack Coho" = "darkorchid4", "Steelhead" = "springgreen4")

# Make seasonplot
fish_ts_longer %>%
  gg_season(y = monthly_mean) +
  labs(x = "Month",
       y = "Mean Number of Fish Observed",
       colour = "Year",
       title = "Seasonal Adult Passage of Coho, Jack Coho, and Steelhead at Willamette Falls") +
  theme_minimal() +
  theme(
    strip.background = element_rect(color = fish_color,
                                    fill = fish_color),
    strip.text = element_text(color = "white")
  )

Figure 1.2 Passage of Coho, Jack Coho, and Steelhead at Willamette Falls for all years between 2001 and 2010, plotted to show the seasonality. Data were shared by and accessed from Columbia River DART (Data Access in Real Time).

Summary

  • The mean counts for Coho and Jack Coho peak during the months of August to October, while the mean count for Steelhead peak around May and June.

  • Coho had the highest peak mean count during the years 2009 and 2010. Jack Coho has the highest peak mean count during the year 2008. Steelhead had higher peak mean counts during the years 2002-2004.

Summary Statistics and Analysis

## Make data frames for each fish species by year
coho_year <- fish_ts %>% 
  index_by(yr = ~year(.)) %>% 
  summarize(year_count_coho = sum(coho, na.rm = TRUE))

jack_year <- fish_ts %>% 
  index_by(yr = ~year(.)) %>% 
  summarize(year_count_jack = sum(jack_coho, na.rm = TRUE))
steel_year <- fish_ts %>% 
  index_by(yr = ~year(.)) %>% 
  summarize(year_count_steel = sum(steelhead, na.rm = TRUE))


jack_year <- fish_ts %>% 
  index_by(yr = ~year(.)) %>% 
  summarize(year_count_jack = sum(jack_coho, na.rm = TRUE))

steel_year <- fish_ts %>% 
  index_by(yr = ~year(.)) %>% 
  summarize(year_count_steel = sum(steelhead, na.rm = TRUE))


## Plot annual totals for fish passage, for each of the 3 species
ggplot() +
  geom_line(data = coho_year,
       aes(x = yr,
           y = year_count_coho,
           colour = "Coho"),
        alpha = 0.7,
        size = 1.2) +
  geom_line(data = jack_year,
       aes(x = yr,
           y = year_count_jack,
           colour = "Jack Coho"),
       alpha = 0.7,
       size = 1.2) +
  geom_line(data = steel_year,
       aes(x = yr,
           y = year_count_steel,
           colour = "Steelhead"),
       alpha = 0.7,
      size = 1.2) +
  scale_x_continuous(breaks= c(2000,2001,2002,2003,2004,2005,2006,2007,2008,2009,2010)) +
  labs(x = "Year",
       y = "Number of Fish",
       colour = "Species",
       title = "Annual Adult Passage of Coho, Jack Coho, and Steelhead at Willamette Falls") +
  theme_minimal() +
  scale_colour_manual(values = c("dodgerblue4", "darkorchid4", "springgreen4"))

Figure 1.3 Annual total Coho, Jack Coho, and Steelhead passage observations were recorded at the Willamette Falls fish ladder on the Willamette River (Oregon) from 2001-2010. Data were shared by and accessed from Columbia River DART (Data Access in Real Time).

Summary

  • From 2001 to 2010, Steelhead observations showed a steady decline, with a small rebound between 2009-2010.

  • Coho observations increased drastically, from ~5,000/year to ~25,000/year in 2009.

  • Jack Coho observations remained relatively consistent over the 10-year period.